|Home | Blog Index | Blog Archives | Christianity & Faith Essays
The El Reno tornado rating downgrade: a temporary product of bureaucracy
"There's no question this was an EF-5 tornado" - Dr. Howie Bluestein
"I don't care what the National Weather Service says, El Reno was an EF-5" - Dr. Chuck Doswell
Those quotes are from two of the top tornado scientists in the world, and are a good representation of the consensus in the community.
Tornado damage ratings have always been, and will likely continue to be, a highly controversial subject in the field of tornado science. This conflict has most recently manifested itself in the rating of the May 31 El Reno, Oklahoma tornado. The El Reno tornado was an EF-5 intensity tornado that was nonetheless recorded in official records by the NWS/NOAA as an EF-3. Both DOW and RaXPol mobile radar units measured near-surface winds of EF-5 strength in the tornado, finding peak velocities of 296MPH. The threshold for EF-5 is 200mph.
Following the tornado's EF-5 rating, NWS leadership issued an internal memo, citing Directive NWSI 10-1604, ordering the cessation of the practice of using mobile radar data to complement damage surveys in the rating of tornadoes. This effectively struck down the original EF-5 rating of the El Reno tornado by the Norman, OK NWS office, leaving the EF-3 damage found along the tornado's path (in the very rural country) as the only criteria permitted to be used in the rating.
When Dr. Ted Fujita introduced the original F-scale in the early 1970s, he was interested in a way to determine and classify the intensity of a tornado. At the time, the only accepted method to estimate this was to observe the damage produced by a tornado, and then infer a windspeed capable of creating that level of damage. In other words, the damage survey was simply a means to an end in attempting to answer the question of how strong a particular tornado was. Likewise, the EF (Enhanced Fujita) scale, implemented in 2007, introduced a more refined and numerous set of "Damage Indicators" (DIs) that are thought to more accurately estimate a tornado's ground-level windspeeds (and thereby its intensity).
The exclusively damage-based nature of both the original F scale and the new EF scale presents one major flaw. It is a long-known problem that the intensity of tornadoes will tend to be greatly underestimated when they occur in rural areas and thus do not strike many 'damage rateable' objects. For example, it's common for strong tornadoes in open country to receive EF/F ratings of 0 or 1, when the same tornado striking a populated area clearly would be capable of much higher rated damage.
For a rural tornado to receive a high rating, it must strike a 'high-end rateable' object at the exact times of its peak strength. The odds of this occuring with any given rural tornado are very low. Most EF4/EF5 tornadoes are ones that have had the "benefit" of striking populated areas, and even then, they only produce their top-level damage in isolated spots. Greensburg and Moore, for example, both had less than 2 instances of EF5 level damage in their entire tracks, both of which impacted broad and dense zones of populated areas containing hundreds of structures (see the map of the Moore, Oklahoma EF-5 track below). If those tornadoes barely attain EF-5 status, you can see, then, the virtual impossibility of an equally-intense tornado in rural areas gaining the proper rating (that reflects its true intensity) from damage alone. It's been generally accepted that there have likely been many, many more F5/EF-5 tornadoes than our current records suggest.
Map of the Moore, Oklahoma EF-5 tornado damage from the Norman, OK National Weather Service, showing the very small zone of EF-5 damage relative to the entire tornado track. Key: EF-0 (light blue), EF-1 (green), EF-2 (yellow), EF-3 (orange), EF-4 (red), and EF-5 (purple)
Source: Norman, OK NWS Office (OUN)
Unlike the use of damage surveys, the use of mobile radar introduces an acceptable way to not just estimate, but measure windspeeds in a tornado. With the mobile radars, the true intensity of a tornado can be determined - getting us closer to the end goal that Fujita was after with his damage-based scale. Using mobile radar, even tornadoes that occur in rural country and strike no high-end-rateable DIs can be discerned as EF-3, 4 or 5 intensities.
The counter-argument is that to maintain consistency in our data, the old system must be used. After all, if we begin counting more rural tornadoes as high-end events with EF4 and EF5 ratings, it won't mesh well with the old data. The problem is that the old data is itself inconsistent and flawed (severely under-rating most rural tornadoes), so any "consistency" achieved would be consistency with poorer data. It's also pointed out that very few tornadoes can be measured with mobile radar, but this is not different from the fact that very few violent rural tornadoes will strike DIs necessary to reflect their true intensity.
The benefit to allowing the mobile radar measurements in ratings is that it will decrease, albeit slightly, the number of under-rated violent tornadoes in our climate data. Since tornado intensity climate data has real-world ramifications (like building codes, perceived risk by the public, etc), society can only benefit by these high-end tornadoes being rated for their true strength.
Now I know that some might say that I have a serious bias in favor the EF-5 rating given my experience on May 31, but I'm truly open to hearing the evidence on both sides. If the tornado was truly of EF-3 intensity, then so be it. What speaks volumes to me is the fact that a significant percentage of the scientific community supports the mobile radar-based ratings and are outspoken against the current NWS stance. This includes many respected figures in tornado science as well as the Norman, OK NWS office that originally rated the tornado. Even the original official recommendation paper for the EF scale states the following: "the technology of portable Doppler radar should also be a part of the EF scale process, either as a direct measurement, when available, or as a means of validating the wind speeds estimated by the experts."
The evidence points to the El Reno tornado being every bit of the record-breaker that we first heard about. The official EF-3 rating appears to simply be a temporary product of bureaucracy, one that will likely be reversed once the influence of the science community is able to infiltrate and/or influence the official policies of the NWS and NOAA. That may take years to happen, however inevitable it may be. In the meantime, we'll just have to accept with a big asterisk the tornado ratings that exclude available mobile radar data under the current policy. By doing so, we all will stay aware of how powerful this tornado truly was, and keep in mind that the next El Reno-like setup will present every bit of the same immense danger to the public and to storm observers as this last one did.
|There needs to be a new rating system that takes into account more than just an estimated wind speed based upon damage. Other factors determine a tornado's power, such as time of day, duration, track width, length, etc. If a new system were to be designed with all of these things taken into account, I think it would help. I for one agree with the downgrade as the EF-Scale is supposed to be about damage, and then accounting for speed based upon that. It's a terrible way to rate a tornado, but I can see why they did it back in the day. It needs scrapped. Many of the factors that I pointed out that I feel need included have been recorded for many tornadoes going back many years. It wouldn't be hard to compare older tornadoes to newer ones using the multi-factor approach.|
- Posted by Jack Frost from Midwest City